Old Franks Fishpasses--1996 Monitoring Update
Prince of Wales Island, Southeast Alaska
John Hannon, Craig Ranger District

Thanks to Kevin Buckley for analyzing the numerous fish scales and for spending lots of time in the field sampling.  Herb Roerick, Jim Zokan, Todd Adams, Pat Moore, and Karen McCartney spent many hours in the field surveying the lake and stream and lived at the fishpass for weeks at a time.  Without the dedication and hard work of these folks this project would not have been possible.

Note:  some figures and tables are distorted in some browsers from conversion to HTML.  The main Arcview map identifying sampling locations is not included.  Contact the author if you want a hard copy of the report.

Monitoring continued on the Old Franks system in 1996 and focused on determining progress toward meeting project objectives and addressing issues identified during environmental analysis for the project. The main objective of the project was to increase the number of salmon, primarily coho and sockeye, available to the common property fishery. The following were issues associated with the project: 1) mixing of Old Franks enhanced stocks with other wild stocks in the commercial fishery; 2) genetic concerns over the stock used for bioenhancement; 3) the effect of anadromous introductions on resident fish in Old Franks; and 4) the effect of additional timber harvest in the watershed on the project. The 1996 monitoring activities included the following:

o Coded wire tag collection from the commercial and sport fisheries

o Escapement monitoring at the upper fishpass and in selected parts of the stream system

o Fish surveys in Old Franks Lake, Lake Mary, and Trail Creek

o Snorkel population surveys in upper Old Franks Creek

o Stream survey of Trail Creek and Toque Creek

o Water temperature and streamflow monitoring

Background Information

The Alaska Department of Fish and Game began fisheries investigations in the Old Franks Watershed in 1978. These were augmented with additional fisheries data collected in 1989 (Zadina and Haddix 1993). The Forestry Sciences Lab collected baseline population and habitat data on the system in 1990 to evaluate the response of the system to the introduction of anadromous salmonids (Bryant 1991). The Craig Ranger District began to monitor population and habitat conditions in the system in 1991 (Craig Ranger District 1994 and 1995).

Two fishpasses were constructed over waterfalls on lower Old Franks Creek in 1992. Coho salmon (Oncorhynchus kisutch) and pink salmon (O. gorbuscha) negotiated the fishpasses during the 1992 run. Coho moved into the upper reaches of the accessible habitat during that year while pink salmon did not travel all the way to the upper fishpass. Pink numbers have since increased and some pink salmon now make it through the upper fishpass. Chum salmon (O. Keta) have also begun to colonize the habitat between the fishpasses.

Sockeye salmon (O. nerka) fingerlings were stocked in 1992 and coho fingerlings were stocked in 1994, 1995, and 1996 (Table 1). The coho bioenhancement program was a three year project conducted by the Alaska Department of Fish and Game. Egg takes were conducted above Salmon Lake in McGilvery Creek, in the Karta system. Egg takes were conducted in late October and November. Sockeye and king salmon were both stocked into the system in the past. No sockeye or king salmon runs developed, but there is a kokanee population in the Old Franks Lake system which may have been established from the sockeye stocking. Presently no other stocking is planned. Resident salmonids present above the barrier waterfalls in the system prior to fishpass construction and stocking were cutthroat trout (O. clarki), rainbow trout (O. mykiss), and dolly varden char (Salvelinus malma).

Table 1. Stocking History of the Old Franks system

Date	Species	Age	Number	Location Planted	Source
1952	Sockeye	Green Eggs	60,000	Unknown	Buschmann Creek
1952	Sockeye	Eyed Eggs	35,000	Unknown	Buschmann Creek
1953	Sockeye	Green Eggs	165,550	Unknown	Buschmann Creek
1962	King Salmon	Fry	33,750	Lakes	Soos Cr., WA
1963	King Salmon	Fry	46,223	Lakes	Karta (Beaver Falls reared)
5-14-92	Sockeye	Fry	227,200	Lakes	Karta (Klawock H reared)
8-16-94	Coho	Fry	96,632	Upper Lake	Karta (Klawock H reared)
7-21-95	Coho	Fry	78,719	Bridge	Karta (Klawock H reared)
7-26-95	Coho	Fry	139,405*	O.F. Lake	Karta (Klawock H reared)
6-25-96	Coho	Fry	143,416	O.F. Lake	Karta (Beaver Falls reared)

34,735 cohos were coded wire tagged in 1994, 32,728 were tagged in 1995, and 44,431 were tagged in 1996.

*9,223 cohos with the same tag code were accidentally released into the Klawock River in 1995.

Methods

Site Description

The Old Franks watershed on Prince of Wales Island is 32 km east of Craig (Figure 1). It covers 65 km² and flows into Polk Inlet (Figure 2). The lakes in the system cover over 300 ha and there are over 19 km of anadromous streams made accessible by the fishpasses. Monitoring work has been conducted primarily on the mainstem of Old Franks Creek below the lakes (escapement), above the lakes (juvenile fish) and within the lakes (resident fish).
 

Escapement Counts

A Smith Root Model 1100 Fish Counter was installed in a horizontal flume at the upper fishpass and has been used to count returning adult salmon since 1993. The picket fish trap was installed in May of 1996 in the exit pool of the fishpass, above the fish counter. The trap helped to validate counts made on the electrical fish counter. Adult salmon captured in the picket trap were measured (mid-eye to fork length), examined for adipose clips, and tagged with a 30 cm yellow spaghetti tag. The tag allowed us to identify fish upstream that passed through the fishpass and to identify fish that went back down over the falls. We took scales from up to ten coho salmon each day and from all sockeye and pink salmon. Fifteen of the adipose-clipped coho were killed to recover the coded-wire tags. The tags identify which stocked group the fish is from. We also conducted periodic snorkel counts from Lake Mary down to saltwater to monitor the upstream migration of the salmon. A Peterson mark-recapture estimate (Chapman modification) was used to estimate escapement through both fishpasses. Confidence intervals were calculated using Appendix II of Ricker (1975).
 

Lake Fish Surveys

Lake fish populations were estimated with Schumacher Eschmeyer and Schnabel mark recapture estimates. Population surveys were conducted between May and August in Old Franks Lake and Lake Mary. We captured fish in fyke nets and large minnow traps. All fish captured were measured to the nearest millimeter (fork length) and weighed to the nearest gram. We took scale samples from five fish of each species in each 20 mm size class in each lake (198 cutthroat and 131 rainbow). Scales were arranged on acetate slides and magnified 48X with a Cannon Micro Printer 90 Microfische reader (Nielson and Johnson, 1983). Scales were read twice. When two readings did not match, a third reading was taken. If a third age was determined the sample was thrown out. Ages were determined according to Pearcy et.al. (1990). Each fish captured was marked on the upper lobe of the caudal fin with an alcian blue dye mark applied with a high pressure (Madajet) injector (Thedinga and Johnson 1995). Recaptured fish were marked a second time on the lower lobe of the caudal fin. We also counted and noted the presence or absence of adipose fins on all coho caught. Coho population estimates were not attempted because both smolting and additional coho stocking occurred during the sampling periods.
 

Trail Creek Fish Survey

We estimated the juvenile coho population in August in Trail Creek, upstream of the lake, using a Schumacher Eschmeyer estimate. Fish were captured in minnow traps baited with salmon eggs. All fish captured (coho, cutthroat, and dolly varden) were measured and dye marked. Insufficient numbers of cutthroat or dolly varden were caught to conduct population estimates on those species.
 

Stream Population and Habitat Surveys

Habitat surveys were conducted in 1990, 1991, 1992, and 1994 on upper Old Franks Creek from the upper lake to the top of anadromous fish habitat. Fish densities (all ages and species) were determined annually since 1990 by snorkeling established habitat units. New snorkel units were marked periodically to adjust for natural changes in habitat units. The draft Region 10 stream survey protocol (Coghill 1996) was used in 1996 to survey Trail Creek and part of Toque Creek.
 

Water Temperature and Level Monitoring

Since 1991, water temperature has been monitored above and below the lake system during summer and fall by using Ryan Temp Mentors recording at one-hour intervals. Water level was recorded daily in lower Old Franks Creek using readings from a staff gauge at the outlet of Lake Mary.

Results

Adult escapement in Old Franks Creek

Coho Salmon

The 1996 coho escapement was the highest yet. It included the first year of returns from fish stocked in 1994. It also included returns from coho that spawned above the fishpass in 1992 and 1993. The picket trap was installed July 8. The first coho was caught soon after installation. Four salmon were seen in the stream reach above the fishpass prior to trap installation, so a few early coho were probably missed. A total of 488 coho were captured in the picket trap by September 30 (Figure 3). Approximately 20 coho went through the fishpass on the night of August 11 when the pickets were removed during rising water. Salmon were also observed jumping out of the picket trap. In addition, during higher flows in late August and September we observed many more coho jumping the falls than going through the fishpass. A section of laminar flow along the side of the fishpass allows coho and sockeye to negotiate the falls.

Figure 3. Old Franks Creek coho salmon run timing through the upper fishpass, 1995 - 1996.
 

Upstream escapement counts were conducted in October to refine the escapement estimate. We conducted a count on October 10 on Trail Creek above the lake. One of 29 coho checked for spaghetti tags was tagged. On October 30 we conducted a count on upper Old Franks Creek from the bridge down to the upper lake. Three of 58 coho checked for tags were tagged. The escapement estimate, assuming a 10% tag loss, is 6,664 coho (95% CI = 2,975 - 16,660). This compares to an escapement estimate of 2,222 (95% CI = 992 - 5,557) in 1995 and approximately 250 in 1993 and 1994 (Figure 4).

Figure 4.  Old Franks Creek coho escapement estimates, 1993 - 1996.  Error bars are 95% confidence intervals.
 

We examined 431 coho at the fishpass for missing adipose fins. Of these 431 coho, 24% had adipose clips. The percentage of tagged coho fingerlings released was 36%. Therefore approximately 66% of the return was from the Karta stock fish and 34% was from Old Franks spawned fish. The Alaska Department of Fish and Game collected 217 Old Franks tagged coho from the commercial fishery. The estimated bioenhancement contribution to the commercial fishery was 1,915 coho. Two sport caught coho were also collected, but no sport contribution was estimated.

The length-frequency distribution of coho at the fishpass showed a peak at 610 mm (mid-eye to fork length), while the distribution of Old Franks coho caught in the commercial fishery (identified by coded wire tags) peaked at 670 mm (Figure 5). The average length of coho that passed through the fishpass was 598 mm (standard deviation = 48). The average length of coho in the commercial fishery was 655 mm (standard deviation = 45).

Figure 5.  Old Franks adult coho, sockeye, and pink salmon length frequency distributions.
 

Sockeye salmon

The 1996 return was the second year of sockeye returns from the 227,000 fingerlings released in 1992. A total of 32 sockeye were caught in the trap at the fishpass (Figure 6). Some sockeye were missed prior to installation of pickets on July 8. Some sockeye also got past the trap on the night of August 11 when the pickets were removed between crew changes. The counter indicated that 58 fish passed through. The assumption is that there were at least 20 sockeye and 20 coho that passed through during that period of rising water. In addition, numerous salmon were observed jumping the falls beside the fishpass. Although none were positively identified as sockeye, it is probable that both sockeye and coho were jumping the falls. Three coded wire tagged sockeye were returned in the commercial fishery. The estimated contribution to the commercial fishery was 105 sockeye. No adipose clipped sockeye were identified at the fishpass.

Figure 6.  Old Franks sockeye salmon run timing through the upper fishpass.
 

Pink and Chum Salmon

We initially believed pink salmon would not make it to the upper fishpass due to cascades between the two fishpasses. Pink salmon escapement above the lower fishpass has increased since the fishpass was built, and in 1995 one pink salmon made it past the upper fishpass. Between August 15 and September 6, 1996 27 pink salmon were caught in the trap at the upper fishpass. Also this was the first year that a chum salmon ascended the upper fishpass, on September 16.
 
 

Lake Fish Populations

Cutthroat

Cutthroat growth (length and weight) in Old Franks Lake was relatively constant through age three and then slowed slightly from age four to six. Growth was relatively constant through age six in Lake Mary (Figure 7, Table 2).

Figure 7.  Cutthroat trout length at age including mean, maximum, and minimum lengths with sample size of fish aged.
 

Table 2. Average fork lengths, weights, and number of fish sampled in each age class.
 

Old Franks Lake cutthroat trout
Age	Two	Three	Four	Five	Six
Average Length (mm)	135	239	289	299	333
Average Weight (g)	31	141	231	248	348
Number of Fish	26	53	41	15	2

 

Lake Mary cutthroat trout
Age	Two	Three	Four	Five	Six
Average Length (mm)	138	216	306	367	440
Average Weight (g)	29	129	361	588	1000
Number of Fish	14	18	21	7	1

 

Old Franks Lake rainbow trout
Age	Two	Three	Four	Five	Six
Average Length (mm)	110	166	218	268	N/A
Average Weight (g)	15	44	98	159	N/A
Number of Fish	18	28	30	11	0

 

Lake Mary rainbow trout
Age	Two	Three	Four	Five	Six
Average Length (mm)	114	181	218	245	214
Average Weight (g)	16	59	183	280	145
Number of Fish	14	13	11	6	1

No fish older than age six were captured in 1996, while fish up to age eight were captured in 1991 and fish up to age eleven were captured in 1978. The cutthroat length frequency distribution in Old Franks Lake was nearly identical to 1995; both had a larger proportion of large fish (300 mm range) than in 1990 (Figure 8). The 1978 distribution had a higher proportion of fish 350mm and larger than any other year surveyed. The cutthroat population was not significantly different than the 1995 estimate, but a much tighter 95% confidence interval of 1223 - 1383 fish (Figure 9) was obtained. Mean length of all cutthroat caught was similar to 1995. The length frequency distribution of cutthroat in Lake Mary was similar to 1995 (Figure 10). The Lake Mary cutthroat population was significantly greater in 1996 than in 1995 (95% confidence interval of 102 - 346 in 1995 and 366 - 434 in 1996) (Figure 11).

Figure 8.  Old Franks Lake cutthroat trout length frequency distribution, 1978, 1991, and 1996.
 

Figure 9.  Population estimates with 95% confidence intervals and mean fork lengths of fish captured in
Old Franks Lake during 1995 and 1996.
 

Figure 10.  Lake Mary cutthroat trout length frequency distribution, 1995 and 1996.
 

Rainbow

Rainbow length showed a relatively linear increase through age five in both lakes, while weights tended to be higher in Lake Mary than in Old Franks Lake (Figure 12, Table 2). Only one rainbow older than age five was captured. The rainbow length frequency distribution in Old Franks Lake had fewer large fish (250 mm+) in 1996 than in 1995, while the 1995 and 1996 distributions both showed a higher proportion of fish in the 200mm+ range than in 1990 (Figure 13). The rainbow population estimate was higher in 1996 (95% confidence interval = 1223 - 1383) but the difference is not significant because the confidence interval was large in 1995 (Figure 9). The rainbow length frequency distribution in Lake Mary also had a smaller proportion of large fish in 1996 than in 1995 (Figure 14). There was a larger proportion of small (<120 mm) rainbow in 1996 than in 1995. There was no significant difference in rainbow population between 1995 and 1996.
 

Figure 11.  Population estimates with 95% confidence interval and mean fork length of fish captured in
Lake Mary during 1995 and 1996.
 
 

Figure 12.  Rainbow trout age length relationship in Old Franks Lake (O.F.L.) and Lake Mary (L.M.), 1996.
Error bars are +/1 one standard deviation.
 
 

Figure 13.  Rainbow trout length frequency distribution in Old Franks Lake, 1990, 1995, and 1996.
 

Figure 14.  Length frequency distributions of rainbow trout in Lake Mary, 1995 and 1996.
 
 
 
 

Potential hybrids

Twenty-one fish caught in Old Franks Lake and four fish caught in Lake Mary had physical characteristics of both cutthroat and rainbow. The maxillary length and coloration of these fish was indeterminate of the two species. These fish were classified as either cutthroat or rainbow and noted as potential hybrids. No genetic analysis was pursued.

Dolly Varden

The dolly varden length frequency distribution in Old Franks Lake had a higher proportion of small fish in 1990 and 1995 than in 1996 (Figure 15). The population estimate was significantly greater in 1996 (95% CI = 5767 - 7274) than in 1995, because the sampling methods using large minnow traps in 1996 were more effective in capturing dolly varden. Thus, the population was more completely sampled (Figure 9). Lake Mary also had more small dolly varden in 1995 than in 1996 (Figure 16), and the confidence interval was tighter in 1996 due to the use of large minnow traps (Figure 11).

Figure 15.  Old Franks Lake Dolly Varden length frequency distribution in 1978, 1990, 1995, and 1996.
 

Figure 16.  Length frequency distributions of dolly varden in Lake Mary, 1995 and 1996.
 
 

Coho

Coho sampling in Old Franks Lake was conducted primarily during smolt outmigration. Therefore, the length frequency distribution for 1996 has a large proportion of smolts, with some smaller holdover or newly emergent coho (Figure 17). Population estimates were not made due to smolting and coho stocking during sampling. Following the first sampling period coho were not measured, but they were counted and checked for the presence of the adipose fin. Prior to coho stocking in 1996 we examined 348 coho. Of those, 13% had clipped adipose fins. Of the coho stocked in 1995, 14% had adipose clips. Following stocking into Old Franks Lake, we examined 937 coho. Of those, 23% had adipose clips. Of the coho stocked in 1996, 28% had adipose clips. Coho were sampled in Lake Mary in July and August, after stocking had occurred in Old Franks Lake. We captured 1,138 coho in Lake Mary and only 3.8% had adipose clips.

Figure 17.  Old Franks Lake coho length frequency distribution in 1995 and 1996.
 

Trail Creek

Coho were the dominant species in Trail Creek. The length frequency distribution (Figure 18) showed that the coho in this small stream grew slower than those in the lake system. The population estimate for coho was 1038 +/- 63 in the 572 meter reach surveyed. Coho density in the stream was 0.23 fish per square meter. One age 0 adipose clipped coho was captured in Trail Creek. The cutthroat and dolly varden captured were larger than the coho. None were age 0 size fish. No cutthroat or dolly varden were recaptured for population estimates.

Figure 18.  Length frequency distributions of coho, cutthroat, and Dolly Varden caught in Trail Creek, 1996.

Salmonid Abundance and Distribution in Upper Old Franks Creek

No coho were stocked into upper Old Franks Creek in 1996; they were all stocked in Old Franks Lake. Cascades above Old Franks Lake prevent upstream migration of juvenile fish. Coho were stocked into upper Old Franks Creek in 1995. The majority of the stocked fish smolt in their first spring due to their larger size at stocking so it is assumed most of the coho in upper Old Franks Creek are Old Franks spawned fish. A snorkel survey of 60 habitat units in upper Old Franks Creek showed coho densities to be lower in slow water habitats and higher in fast water habitats than in 1995 (Figure 19). The 1996 survey is a more natural distribution of habitat usage because coho were not stocked in the stream in 1996. The 1995 survey was conducted only two weeks

following stocking. Coho densities were higher in scour pools (plunge, mid-channel scour, and lateral scour) than in other habitat types (Figure 20). The coho were more evenly distributed throughout the habitat types in 1996 than in 1995, but overall densities were higher in 1995.

Figure 19.  Fish densities in upper Old Franks Creek in slow water and fast water habitats, 1990 - 1996.
 
 

Figure 20.  Cutthroat/rainbow (combined)  and coho densities by habitat type in upper Old Franks Creek, 1996.
pl-bw = backwater pool, pl-ed = eddy pool, pl-gl = glide pool, pl-ls = lateral scour pool, pl-mcs = mid-channel scour pool, pl-pp = plunge pool,
pl-tr = trench pool, rf-gl = riffle glide, rf-rf = riffle, rf-b = boulder riffle, rf-cb = cobble riffle, rf-rp = rapids, All = all habitat types combined.
 

The cutthroat/rainbow density was lower in 1996 than in 1994 or 1995, but still higher than during the period between 1990 and 1993 (Figure 19). The size distribution in 1996 indicated that 30% of the cutthroat/rainbow were age 1+ fish and 70% were age 0 fish. The distribution in 1995 and 1994 indicated that 25% and 23% were age 1+ fish in those years respectively. The cutthroat/rainbow density by habitat type was lower than the coho density in all pool habitats except eddy pools. In fastwater habitats, the cutthroat/rainbow density was higher than coho only in glides (Figure 20). The majority of the cutthroat/rainbow and coho in the stream were located in mid-channel scour pools, lateral scour pools, and riffles (Figure 21). Fish populations extrapolated from densities were 2,261 coho and 1,416 cutthroat/rainbow. These are below actual populations because some fish are missed during snorkel counts.

The dolly varden density was lower in 1996 than in 1991 through 1995, but the density has been consistently very low in upper Old Franks Creek. During 1996, observations revealed dolly varden were more visible and tended to be more active during moderate streamflows than they were during low flows (i.e. they tended to be swimming up in the water column rather than sitting on the bottom of the stream or under rocks).

Figure 21.  Cutthroat/rainbow (combined) and coho populations by habitat type in upper Old Franks Creek, 1996.
pl-bw = backwater pool, pl-ed = eddy pool, pl-gl = glide pool, pl-ls = lateral scour pool, pl-mcs = mid-channel scour pool, pl-pp = plunge pool,
pl-tr = trench pool, rf-gl = riffle glide, rf-rf = riffle, rf-b = boulder riffle, rf-cb = cobble riffle, rf-rp = rapids, All = all habitat types combined.
 

Trail Creek Stream Survey

A tier 3 stream survey was conducted on all anadromous stream habitat on the Trail Creek system. The system includes a 64 meter reach of MM1 channel between Old Franks Creek and a small lake (average bankfull width = 4 meters). Above the lake there are two anadromous forks, east fork and west fork of Trail Creek. The east fork of Trail Creek contains the majority of the anadromous spawning and rearing habitat. The west fork contains 92 meters of anadromous habitat in an MM1 channel (average bankfull width = 3.5 meters). Survey data analysis is presented only for the east fork (Table 3). Comparison with U.S. Forest Service, Alaska Region Riparian Habitat Management Objectives (USDA Forest Service 1995) puts the east fork of Trail Creek at the far end of the ranges for all three parameters in the MM1 channel type. Trail Creek was above the 75th percentile for pieces of large woody debris per 1,000 square meters. It was well above the 75th percentile for percent pool area. It was below the 25th percentile for bankfull width to depth ratio. No Regional Riparian Habitat Management Objectives have been developed for PA5 channels. The habitat in the MM1 channel type consisted primarily of scour pools and sloughs. The habitat in the PA1 channel consisted primarily of slough habitat (Figure 22). Dominant substrate in the MM1 channel consisted primarily of course gravel and in the PA5 channel it consisted primarily of sand (Figure 23).

Figure 22.  Trail Creek habitat types by channel type (BW = backwater pool, G = glide, RF = riffle, SL = slough,
SP = scour pool, MM1 = moderate gradient mixed control channel, PA5 = palustrine channel type).
 
 

Figure 23.  Trail Creek dominant substrates by channel type, 1996.
 

A portion of the upper watershed of Trail Creek was logged in approximately 1991. The logging appears to have had no detectable effect on fish habitat in Trail Creek. Defined stream channels are not present in the logged portion of the watershed. The channels originate primarily in muskeg on a flat gradient. The habitat in Trail Creek is ideal for coho salmon. There is ample spawning habitat for adults and abundant shallow lake and slough habitat (beaver ponds) for coho rearing. Fish population survey data is presented in an earlier section of this report.
 

Table 3. Tier 3 stream survey statistics for the east fork of Trail Creek.
 

C-Type	Length of Channel Type, m	% Pools	Pools/km	BFW/Resid Pool Depth	Average Resid Pool Depth, m	BFW/ Depth (CTV)	LWD #1 /1000 m**2	LWD #2/ 1000 m**2	LWD Key /1000 m**2	LWD Total/ 1000 m**2	Average Bankful Width, m	Undercut length /stream length
MM1	304	79	92	10	.24	8.8,2.24.6	7.1	5.3	47.9	97.5	2.4	.16
PA1	268	96	48	4.8	.58		4.0	0	17.3	34.5	2.2	.31

Water Temperature and Streamflow

The high water temperature for the year occurred on July 25 both below and above the lake system. The high temperature below the lakes was 24.4 degrees celsius, and the high above the lakes, near the bridge crossing, was 15.6 degrees celsius. This was slightly higher than the maximum for 1995 (Figure 24). No fish kills were observed during periods of high water temperature in 1996. Over 100 salmon (coho and sockeye) were observed in lower Old Franks Creek in early July and were present in the stream during the high temperatures in July. The average temperature difference from the bridge to below the lakes is 8 degrees celsius.
 

The majority of returning salmon passed through the fishpass during periods of rising streamflow and periods of a constant moderate flow in lower Old Franks Creek (Figures 25, 3, & 6). During low flow periods the salmon tended to stay in the deepest pools, not moving until the next high water event.

Figure 24.  Water temperatures above (Upper) and below (Lower) the Old Franks lake system, 1995 and 1996.
 

Figure 25.  Water level in lower Old Franks Creek (staff gauge) and rainfall during salmon escapement counts.

Discussion

Salmon Escapement

Coho salmon escapements into the Old Franks system have increased since the fishpasses were installed in 1992. The first increase in escapement occurred in 1995 and was due to natural reproduction in the newly available habitat above the barrier falls. The increased escapement in 1996 was due to both coho bioenhancement from Karta stock and natural reproduction above the barriers. Coded wire tagging of bioenhanced fish has documented the Old Franks contribution to the commercial troll, seine, gillnet and sport fisheries. A period of heavy sport fishing pressure was evident in mid-August at the mouth of Old Franks Creek. The coho production capability of the system was predicted to be 6,189 coho. It appears that production, with the aid of bioenhancement, is approaching this number.
 

Coho juveniles are present in nearly all accessible habitat below the bridge on upper Old Franks Creek. Few juveniles have been found above the bridge and no adults have been observed above the bridge. The majority of the coho seem to be spawning in the small tributary streams to the lakes and in the streams between the lakes rather than in the extensive habitat in upper Old Franks Creek. We counted only 58 coho in upper Old Franks during an escapement count on October 30.
 

The coded wire tagged coho from Old Franks caught in the commercial fishery were larger than those that passed through the fishpass. The larger size was consistent across all survey locations. The reason for the size difference is unknown. It could be that the larger fish are more vulnerable to capture in the fishery. Another possible explanation is that the larger coho are more powerful jumpers and are able to jump the falls at Old Franks, while the smaller ones get counted and measured because they use the fishpass.
 

The sockeye escapement was similar to 1995, and may have been slightly higher since the pickets were removed on a day of expected peak migration for sockeye. The sockeye return is from the sockeye released into the lakes in 1992. Sockeye returns have not been high enough to fully seed the system. Although no sockeye were released after 1992, sockeye adults will probably continue to return in low numbers as some of the kokanee offspring outmigrate and return as adults.
 

Pink and chum salmon escapement above the lower fishpass has shown a steady increase following fishpass installation. Passage for pink and chum salmon through the cascade falls between the fishpasses is flow dependent, but as escapement increases more pink and chum may make it past the upper fishpass. They will probably not make it past Old Franks Lake due to the falls just above the lake.
 

The fish counter performed similarly in 1995 and 1996. The actual number of fish captured was 60% and 57% of the number counted by the fish counter in 1995 and 1996 respectively. A calibration ratio based on past experience can be used in the future when the fishpass is unstaffed.
 

Resident Fish

Rainbow trout numbers in the lakes are consistently higher than cutthroat numbers. However, the cutthroat become larger in the lakes than the rainbow. Nearly all fish over 300 mm are cutthroat. The cutthroat show a modal length distribution peaking at around 300 mm while the rainbow show more of a bimodal distribution. The Lake Mary cutthroat population was greater in 1996 than in 1995, and the length frequency distributions in Old Franks Lake and Lake Mary showed larger cutthroat in 1996 than in 1995. The size difference may be due to the increased availability of prey, primarily coho juveniles. Although no formal stomach content analysis was conducted, many of the cutthroat caught in fyke nets had distended stomachs full of fish that matched the size of the juvenile coho. Some of the large cutthroat were recaptured up to four times, and each time they had full stomachs. They may have learned that the coho in the nets were easy prey. The cutthroat in Lake Mary appear to grow (length and weight) faster from age four to six than the cutthroat in Old Franks Lake. This may be due to availablity of more coho as prey as they pass through the relatively narrow Lake Mary during outmigration. The number of cutthroat drops off above age three in Old Franks Lake. It is possible that cutthroat are smolting after age three and becoming anadromous. This happened on a similar system with a fishpass on the Cordova Ranger District (Cordova Ranger District 1996). High numbers of returning anadromous cutthroat have not been documented in Old Franks.
 

There was a smaller proportion of large rainbow in both lakes in 1996 than in 1995. The reason for this is unknown. The cutthroat may be more efficient predators on coho than the rainbow are.

Few cutthroat, rainbow, or dolly varden were captured in the survey of Trail Creek, but the ones that were captured were larger than the coho. We had expected that more resident fish would be found rearing in Trail Creek. During fyke netting in the lake on Trail Creek in 1995, few resident fish were captured as well. It appears that this sub-watershed is being utilized more by coho than by resident species. Comparison with regional habitat management objectives indicates the habitat is in good condition.
 

Coho are now the dominant species in upper Old Franks Creek. They were present in most habitats below the bridge crossing. The 1996 cutthroat/rainbow density was lower than in 1994 or 1995 but still higher than in prior years. It appears that the mainstem of upper Old Franks Creek is used little by rearing cutthroat or rainbow. Most juvenile cutthroat and rainbow move out of the mainstem before they are one year old. The majority probably go into the smaller tributaries and some may enter the lakes at that point. More surveys of off-channel habitats are needed to determine where they go. The primary area of habitat use overlap between coho and cutthroat/rainbow in upper Old Franks Creek appears to be in mid-channel scour and lateral scour pools.
 

Acknowledgements
Thanks to Kevin Buckley for analyzing the many fish scales.  Herb Roerick, Jim Zokan, Todd Adams, Pat Moore, and Karen McCartney spent many days and nights in the field under inclement conditions surveying the lake and stream and lived at the fishpass for weeks at a time.  This work would not have been possible without the hard work and dedication of these folks.
 
 
 

Literature Cited

Bryant, M.D.  1991. Evaluation of the zooplankton and resident salmonid populations of Old Franks Lake before the introduction of an anadromous salmonid population. Final Report for 1990. FSL Juneau. 21p.
 

Coghill, K. 1996. Draft stream survey protocol for the Tongass National Forest. June 21, 1996 version. FSL Juneau. 26p.
 

Hannon, J.  1995. Old Franks Monitoring--1995. Unpublished Document. USFS Craig. 25p.
 

Nielson, L.A. and D.L. Johnson eds. 1983. Fisheries techniques. American Fisheries Society. Bethesda, MD. 468p.
 

Pearcy, W.G., R.D. Brodeur and J.P. Fisher. 1990. Distribution and biology of juvenile cutthroat trout Oncorhynchus clarki clarki and Steelhead O. mykiss in coastal waters off Oregon and Washington. Fisheries Bulletin. 88:697-711.
 

Ricker, W.E. 1975. Computation and interpretation of biological statistics of fish populations. Bulletin of the Fisheries Research Board of Canada. Bulletin 191., Ottawa, Canada. 382p.
 

Thedinga, J.F. and S.W. Johnson. 1995. Retention of jet-injected marks on juvenile coho and sockeye salmon. Transactions of the American Fisheries Society 124:782-785.
 

USDA Forest Service. 1995. Report to Congress--Anadromous Fish Habitat Assessment. Pacific Northwest Research Station, Alaska Region. R10-MB-270.
 

Zadina and Haddix. 1993. Summary of limnological and fisheries investigations of the Old Franks Lake system 1978-1989. Alaska Department of Fish and Game FRED Division. 37p.